AI Paper Atlas

Research map / 3D vision / Depth and stereo

REON-NVS: Real-Time Online Novel-View Synthesis from Sparse-View Videos

Daeyeon Kim, Jinhyeok Kim, Gangmin Kwon, Seungjoo Shin, Sunghyun Cho

ECCV 2026 · 3D vision › Depth and stereo · approach: Feed-forward 3D reconstruction · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add flicker loss (ablation)REON-NVS · Neural 3D Video (6 scenes, 2 input views)PSNR: 23.41 -> 23.63Improved
add flicker loss (ablation)REON-NVS · Neural 3D Video (6 scenes, 2 input views)Flicker: 10.27 -> 8.97Improved
add flicker loss (ablation)REON-NVS · Multi-view dynamic scene dataset (30 test scenes)PSNR: 23.09 -> 23.26Improved
add flicker loss (ablation)REON-NVS · Multi-view dynamic scene dataset (30 test scenes)Flicker: 46.30 -> 45.45Improved
add flicker loss (ablation)REON-NVS · RealEstate10K (pixelSplat test split)PSNR: 26.67 -> 27.22Improved
add flicker loss (ablation)REON-NVS · RealEstate10K (pixelSplat test split)Flicker: 20.96 -> 19.91Improved
enable SSM hidden-state temporal memory (ablation)REON-NVS · Retrieval task (revisited regions)PSNRImproved
use bidirectional scanning instead of unidirectional (ablation)Bi-SSM history fusion · RealEstate10K (pixelSplat test split), history fusion onlyPSNR: 19.79 -> 20.82Improved
alternate scan direction across blocks (cross) and use add token merge (ablation)Bi-SSM history fusion · RealEstate10K (pixelSplat test split), history fusion onlyPSNR: 21.08 -> 22.04Improved
replace RayZer's latent compression with direct input images plus SSM temporal memoryREON-NVS vs RayZer · Neural 3D Video (6 scenes, 2 input views)PSNR: 21.97 -> 23.63Improved
replace RayZer with REON-NVS on moving camerasREON-NVS vs RayZer · Multi-view dynamic scene dataset (30 test scenes)PSNR: 21.57 -> 23.26Improved
geometry-free feedforward instead of DepthSplat with ground-truth posesREON-NVS vs DepthSplat · Neural 3D Video (6 scenes, 2 input views)PSNR: 19.81 -> 23.63Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 4 of 4 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy